Study On Wheat Ecological Holographic System

Wheat production has developed from high yield to high yield and high quality.To realize the wisdom of wheat production,to achieve fine seed selection,precise fertilization,precise cultivation,intelligent management and scientific decision-making,which is a major need for the implementation of new technology inheritance and the promotion of agricultural modernization.At the same time,it is also an important means of rational utilization of resources,scientific input,cost saving and efficiency increasing,yield increasing and quality preserving,and pollution reduction,which is a great significance to guarantee national food security.This study,the production process of wheat as a whole ecological information system,the whole process of data was collected in wheat production.Using the crop model,the Internet of things,big data,machine learning and data visualization technology,wheat varieties were studied in different parts of the ecological adaptability,carried out the wheat automatic and semiautomatic production data acquisition,builded the wheat ecological holographic model and system.The main results and conclusions of the research are as follows:1.Established the standard of wheat ecological factor library and the evaluation method of wheat variety suitability.Combined with the wheat production process,30 biological factors and artificial factors affecting wheat production was determined,a variety suitability evaluation method based on crop models was constructed,and the adaptability of 4 varieties in 3 regions was evaluated.The evaluation results showed that: AK58 has the best adaptability in Luohe and Changge,with an adaptability index above 0.8,XN509 has an adaptability above 0.9 in all three regions,and YM49-198 has the best adaptability in Luohe with an adaptability index.Above 0.93,ZM27 has the best adaptability in Luohe and Huaxian,and the adaptability index is above 0.73,indicating that it is effective to use crop models to evaluate the adaptability of wheat varieties.2.The recognition algorithm and model of the number of basic seedling and ear at the seedling stage and mature stage of wheat were established.Based on AR technology,the image of wheat basic seedling can be obtained quickly,and the AR measurement accuracy can reach 99.8%.Using image processing technology to achieve rapid monitoring of the wheat seedling stage,the recognition accuracy of the basic seedlings at the 1~4 leaf stage of wheat is better than 95%.Based on the smart phone,the use of machine learning and migration learning technology has exploited the fast identification and counting algorithm of the wheat ears at the mature stage,and the number of wheat ears identified by the algorithm and the number of wheat ears manually counted R2 are above 0.95,indicating that it is feasible to use AR and image processing technology can be used to recognize wheat seedlings,and it is feasible to use smart phones and migration learning technology to realize rapid identification and counting of wheat ears.3.Established a distributed wheat production management holographic big data warehouse integrating big data storage,processing,calculation,statistics,analysis and visualization.Utilizing web crawler and Internet of Things technology to realize the automatic acquisition and storage of wheat production knowledge and Internet of Things data.Based on the Hadoop ecosystem and using big data,My SQL cluster storage and management of structured data of wheat production is realized,and an integrated storage and calculation model of unstructured data of wheat production is built.Compared with the native HDFS,the read-write efficiency of the storage model of this study is improved by 64.40% and 38.98%,respectively,indicating that it is feasible to use the storage model in this study to store mass data of wheat production.4.Constructed of five types of holographic models covering the whole process of wheat production,including variety selection,precise fertilization,sowing date recommendation,natural disaster monitoring and early warning,growth monitoring and yield prediction.Comprehensive use of model technology,"3S",Internet of Things,machine learning and other technologies,based on the six dimensions of yield,resistance,key cultivation techniques,suitable areas,sowing date,and characteristic characteristics,developed a multi-objective decision-making model for wheat varieties.A multi-dimensional and multi-objective fuzzy optimization decision-making algorithm was used to develop a precise fertilization decision-making model.On this basis,the accumulated temperature theory was used to realize the calculation and recommendation of the sowing amount.Based on the meteorological and Internet of things data,combined with dry hot wind and late frost damage indicators,the natural disaster warning model of dry hot wind and late frost damage of wheat was constructed.Based on crop model and remote sensing assimilation technology,a holographic monitoring model of wheat growth and yield was constructed,and the results showed that using remote sensing and crop model assimilation technology could effectively improve the simulation effect of the model.5.A holographic system of wheat production management covering the whole process of wheat growth and development was developed.Based on micro service technology,combined with large data warehouse model for wheat and wheat production,the holographic wheat production management system,realized the wheat varieties of sowing date and sowing quantity recommended fertilizer,information release rate prediction monitoring meteorological disaster warning and production knowledge visualization,implements the wheat production to the holographic data monitoring and visualization.